import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
import seaborn as sns
import plotly.express as px
import plotly.figure_factory as ff
import plotly.graph_objects as go
from scipy.stats import norm
import statistics as sta
import scipy.stats as stats
import json

Read data

#read data
def result(year):
    path = "data/diemthi"+str(year)+".csv"     #path: sửa cái này để đọc data
    data = pd.read_csv(path, encoding = "Latin-1")
    data ['year'] = year
    return data
data = result(2021)
data

	Unnamed: 0	ID	CityCode	CityArea	Math	Literature	English	Physics	Chemistry	Biology	...	Civic Education	A	B	C	A1	D1	A-BK	A1-BK	Highest_combi	year
0	0	1000001	1	1.0	2.2	3.50	NaN	NaN	NaN	NaN	...	NaN	NaN	NaN	11.50	NaN	NaN	NaN	NaN	C	2021
1	1	1000002	1	1.0	9.2	8.25	10.0	NaN	NaN	NaN	...	9.25	NaN	NaN	23.75	NaN	27.45	NaN	NaN	D1	2021
2	2	1000003	1	1.0	4.4	6.25	NaN	NaN	NaN	NaN	...	NaN	NaN	NaN	14.50	NaN	NaN	NaN	NaN	C	2021
3	3	1000004	1	1.0	8.4	6.75	10.0	4.00	3.00	3.50	...	NaN	15.40	14.90	NaN	22.40	25.15	17.8500	23.1000	D1	2021
4	4	1000005	1	1.0	8.8	8.25	9.4	8.25	5.75	5.25	...	NaN	22.80	19.80	NaN	26.45	26.45	23.7000	26.4375	D1	2021
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
1002265	1021097	64006834	64	6.0	8.6	8.25	6.2	8.25	7.00	6.50	...	NaN	23.85	22.10	NaN	23.05	23.05	24.3375	23.7375	A	2021
1002266	1021098	64006835	64	6.0	9.0	7.50	9.4	8.00	8.25	5.50	...	NaN	25.25	22.75	NaN	26.40	25.90	25.6875	26.5500	A1	2021
1002267	1021099	64006836	64	6.0	4.8	7.00	4.6	NaN	NaN	NaN	...	9.00	NaN	NaN	17.75	NaN	16.40	NaN	NaN	C	2021
1002268	1021100	64006837	64	6.0	7.4	7.75	9.4	NaN	NaN	NaN	...	9.25	NaN	NaN	19.00	NaN	24.55	NaN	NaN	D1	2021
1002269	1021101	64006838	64	6.0	8.0	7.75	8.2	NaN	NaN	NaN	...	7.50	NaN	NaN	18.50	NaN	23.95	NaN	NaN	D1	2021

1002270 rows × 22 columns

City = {1: "Hanoi", 2: "Ho Chi Minh City", 3: "Hai Phong",
        4: "Da Nang", 5: "Ha Giang", 6: "Cao Bang",
        7: "Lai Chau", 8: "Lao Cai", 9: "Tuyen Quang",
        10:"Lang Son", 11:"Bac Kan", 12:"Thai Nguyen",
        13:"Yen Bai", 14:"Son La", 15:"Phu Tho",
        16:"Vinh Phuc", 17:"Quang Ninh", 18:"Bac Giang",
        19:"Bac Ninh", 21:"Hai Duong",
        22:"Hung Yen", 23:"Hoa Binh", 24:"Ha Nam",
        25:"Nam Dinh", 26:"Thai Binh", 27:"Ninh Binh",
        28:"Thanh Hoa", 29:"Nghe An", 30:"Ha Tinh",
        31:"Quang Binh", 32:"Quang Tri", 33:"Thua Thien - Hue",
        34:"Quang Nam", 35:"Quang Ngai", 36:"Kon Tum",
        37:"Binh Dinh", 38:"Gia Lai", 39:"Phu Yen",
        40:"Dak Lak", 41:"Khanh Hoa", 42:"Lam Dong",
        43:"Binh Phuoc", 44:"Binh Duong", 45:"Ninh Thuan",
        46:"Tay Ninh", 47:"Binh Thuan", 48:"Dong Nai",
        49:"Long An", 50:"Dong Thap", 51:"An Giang",
        52:"Ba Ria - Vung Tau", 53:"Tien Giang", 54:"Kien Giang",
        55:"Can Tho", 56:"Ben Tre", 57:"Vinh Long",
        58:"Tra Vinh", 59:"Soc Trang", 60:"Bac Lieu",
        61:"Ca Mau", 62:"Dien Bien", 63:"Dak Nong", 64:"Hau Giang"}

AreaCode = {1: "Red River Delta", 2: "Northern midlands and mountain areas", 3:"North Central and Central coastal areas",
        4: "Central Highlands",
        5: "South East", 6: "Mekong River Delta"}

Area = {1:[1,3,16,19,22,21,26,25,24,27,17],
        2:[5,9,15,12,11,6,10,18,8,13,7,14,23,62],
        3:[28,29,30,31,32,33,4,34,35,37,39,41,45,47],
        4:[36,40,38,42,63],
        5:[2,52,44,43,46,48],
        6:[55,49,53,56,57,58,50,51,54,64,59,60,61]} 

Candidate number of each province

def count_province(data):
    
    A = data['CityCode'].value_counts().sort_index()
    AreaSet = []
    for i in City.keys():
        for j in Area:
            if i in Area[j]:
                AreaSet.append(j)
    AreaName = [AreaCode[i] for i in AreaSet]
    CountTinh = {'CityCode':pd.Series(City.keys()),
                 'Province':pd.Series(City.values()),
                 'AreaCode': pd.Series(AreaSet),
                 'Area':pd.Series(AreaName),
                 'Count':pd.Series(list(A))}
    df = pd.DataFrame(CountTinh)
    fig = px.bar(df, x="Province", y="Count",barmode='group',color='Area',width=1000,height=600)
    fig.update_traces(width = 1)
    return fig

count_province(data)

Candidate number based on highest subject combination

def count_combi(data):
    
    A = data['Highest_combi'].value_counts().sort_index()
    CountTinh = {'Highest_combi':pd.Series(['A','A1','B','C','D1']),
                 'Count':pd.Series(list(A))}
    df = pd.DataFrame(CountTinh)
    fig = px.bar(df, x='Highest_combi', y='Count', labels={'Count':'Participant number'})
    return fig

count_combi(data)

A = data[data.Highest_combi == 'D1']
B = A[A.Physics >= 0]
len(B)/len(A)

0.3164123114702008

Score spectrum

Whole country

def score_spectrum(data, subject):
    subject_name = ''
    if (len(subject)<3 or ('-' in subject)):
        bins = [i/2 for i in range(0,61)]
        data_show = data[subject].value_counts(bins = bins).sort_index()
        subject_name = 'Combination: '+subject
    else:
        data_show = data[subject].value_counts().sort_index()
        subject_name = subject
        
    count_score = {'Score': pd.Series((data_show.index).astype('str')),
               'Count': pd.Series(data_show.values)}
    
    df = pd.DataFrame(count_score)
    fig = px.bar(df, x='Score', y='Count',title=subject_name)
    return fig

score_spectrum(data,'D1')

City and Province

def score_spectrum_province(data,subject,CityCode):

    data_province = data[data.CityCode == CityCode]
    subject_name = ''
    if (len(subject)<3 or ('-' in subject)):
        bins = [i/2 for i in range(0,61)]
        data_show = data_province[subject].value_counts(bins = bins).sort_index()
        subject_name = 'Combination: '+subject
    else:
        data_show = data_province[subject].value_counts().sort_index()
        subject_name = subject
        
    count_score = {'Score': pd.Series((data_show.index).astype('str')),
               'Count': pd.Series(data_show.values)}
    
    df = pd.DataFrame(count_score)
    fig = px.bar(df, x='Score', y='Count',title=City[CityCode]+': '+subject_name)
    return fig

score_spectrum_province(data,'English',1)

Area

def score_spectrum_area(data,subject,CodeArea):
 
    data_province = data[data.CityArea == CodeArea]
    subject_name = ''
    if (len(subject)<3 or ('-' in subject)):
        bins = [i/2 for i in range(0,61)]
        data_show = data_province[subject].value_counts(bins = bins).sort_index()
        subject_name = 'Combination: '+subject
    else:
        data_show = data_province[subject].value_counts().sort_index()
        subject_name = subject
        
    count_score = {'Score': pd.Series((data_show.index).astype('str')),
               'Count': pd.Series(data_show.values)}
    
    df = pd.DataFrame(count_score)
    fig = px.bar(df, x='Score', y='Count',title=AreaCode[CodeArea]+': '+subject_name)
    return fig

score_spectrum_area(data,'English',3)

Whole country score spectrum and Province score spectrum

import matplotlib.pyplot as plt
import seaborn as sns

def Spectrum_Compare(df, CityCode, mon_hoc):
    sns.set_theme()
    p1 = sns.kdeplot(df[mon_hoc], 
             color = 'darkblue', bw_adjust=2, legend = True, cut = 1)
        
    p2 = sns.kdeplot(df[df.CityCode==CityCode][mon_hoc], 
             color = 'red', bw_adjust=2, cut = 1)
    
    plt.legend(labels=['Toàn quốc', City[CityCode]])
    # Khai báo tiêu đề cho trục x
    plt.xlabel('Exam result: '+mon_hoc)
    # Khai báo tiêu đề cho trục y
    plt.ylabel('Exam result density distribution')
    plt.rcParams['figure.figsize'] = (20, 7)

    plt.show()

Spectrum_Compare(data, 28, 'English')

def DistPlotly(year,subject):
    data = result(year)
    A = list((data[data[subject]>=0][subject]).sample(1000))
    hist_data = [A]
    group_labels = [subject] # name of the dataset

    fig = ff.create_distplot(hist_data, group_labels)
    fig.show()

DistPlotly(2021,'English')

sns.catplot(data = data.loc[:,data.columns[4:7]], kind = 'boxen',dodge=False)
sns.catplot(data = data.loc[:,data.columns[7:10]], kind = 'violin',dodge=False)
sns.catplot(data = data.loc[:,data.columns[10:13]], kind = 'box',dodge=False)
sns.catplot(data = data.loc[:,data.columns[13:20]], kind = 'box',dodge=False)
plt.show()

def boxplot(subject):
    df1 = data[data[subject]>=0].sample(1000)
    fig = px.box(df1, y=subject, points="all")
    return fig

boxplot('English')

Area score spectrum

import matplotlib.pyplot as plt
import seaborn as sns

def SoSanhVung(data, AreaSet, mon_hoc):
    sns.set_theme()
    p1 = sns.kdeplot(data[mon_hoc], 
             color = 'darkblue', bw_adjust=2, legend = True, cut = 1)
    for i in AreaSet:
        sns.kdeplot(data[data.CityArea==i][mon_hoc], bw_adjust=2, cut = 1)
    
    plt.legend(labels=['Whole country']+[AreaCode[i] for i in AreaSet])
    # Khai báo tiêu đề cho trục x
    plt.xlabel('Exam result: '+mon_hoc)
    # Khai báo tiêu đề cho trục y
    plt.ylabel('Exam result density distribution')
    plt.rcParams['figure.figsize'] = (20, 7)
    
    plt.show()

for i in data.columns[4:20]:
    SoSanhVung(data, [1,2,3,4,5,6], i)

Subject score spectrum

def SoSanhPhoMon(data, mon_hoc):
    for i in mon_hoc:
        p1 = sns.kdeplot(data[i], 
             bw_adjust=2, 
             legend = True,
             color=mon_hoc[i],
             cut = 1)
    
    plt.legend(labels=mon_hoc)
    # Khai báo tiêu đề cho trục x
    plt.xlabel('Exam result:')
    # Khai báo tiêu đề cho trục y
    plt.ylabel('Exam result density distribution')
    plt.show()

mon_hoc = {"Math": "#45a39b", 
           "Literature": "#a0a345", 
           "English":"#77a345" , 
           "Physics":"#4577a3", 
           "Chemistry":"#5145a3", 
           "Biology": "#8845a3", 
           "Geography":"#a34845", 
           "History":"#a36a45", 
           "Civic Education":"#a3456a"}
SoSanhPhoMon(data,mon_hoc)

score spectrum year to year

def SoSanhPhoNam(year, mon_hoc):
    for i in year:
        data1 = result(i)

        p1 = sns.kdeplot(data1[mon_hoc], 
             bw_adjust=2,
             legend = True,
             cut = 1)
    
    plt.legend(labels=year)
    # Khai báo tiêu đề cho trục x
    plt.xlabel('Exam result: '+ mon_hoc)
    # Khai báo tiêu đề cho trục y
    plt.ylabel('Exam result density distribution')
    plt.savefig(mon_hoc+'.png')
    plt.show()

for i in data.columns[4:20]:
    SoSanhPhoNam([2018,2019,2020,2021], i)

path = "D:/pythonProject/DS capstone project/data/Province_Avg2021.csv"
Province_Avg_Score = pd.read_csv(path, encoding = "Latin-1")

Scatter plot

Combo_data = {'A': data[data.Highest_combi == "A"],
'A1': data[data.Highest_combi == "A1"],
'B': data[data.Highest_combi == "B"],
'C': data[data.Highest_combi == "C"],
'D1': data[data.Highest_combi == "D1"]}

Combo = {"A":["Math", "Physics", "Chemistry"],
         "B":["Math", "Chemistry", "Biology"],
         "C":["Literature","History","Geography"],
         "D1":["Math","Literature","English"],
         "A1":["Math","English","Physics"]
}

def Scatter(Combo, subject):
    X = list(Combo_data[Combo][subject])
    Y = list(Combo_data[Combo][Combo])
    
    fig = px.scatter(data_frame=Combo_data[Combo],x=subject,y=Combo,width=600,height=600)
    
    fig.show()
    print('Pearson correlation:',stats.pearsonr(X,Y)[0])
    print('Spearman correlation:',stats.spearmanr(X,Y)[0])

Scatter('A1','Math')

Pearson correlation: 0.8180867949870909
Spearman correlation: 0.7870674707312136

def HighLow(Combi,mini,subject):
    A = data[data['Highest_combi'] == Combi]
    B = A[A[Combi]>=mini]

    high = len(B[B[subject]>(B[Combi]/3+0.5)])
    low = len(B[B[subject]<(B[Combi]/3-0.5)])
    med = len(B)-high-low
    
    return [high, med, low]

def HighLowDict(Combi,mini):
    df = {'Subject':[Combo[Combi][0]]*3+[Combo[Combi][1]]*3+[Combo[Combi][2]]*3,
          'Type': ['High','Medium','Low']*3,
          'Count': HighLow(Combi,mini,Combo[Combi][0])+HighLow(Combi,mini,Combo[Combi][1])+HighLow(Combi,mini,Combo[Combi][2])
    }
    return pd.DataFrame(df)

def HighLowFig(Combi, mini):
    df = HighLowDict(Combi,mini)
    fig = px.bar(df, x="Subject", y="Count", color="Type", title='Combination: '+Combi)
    fig.show()

''' The plot shows the affect of subject to combination:
High: the number of score of subject is greater than one-third of combination score plus 0.5
Low: the number of score of subject is greater than one-third of combination score minus 0.5
Medium: the remain number of score'''

HighLowFig('A',20)

Chart of each year

bigData = pd.concat([result(2021),result(2020),result(2019),result(2018)])

bigData

	Unnamed: 0	ID	CityCode	CityArea	Math	Literature	English	Physics	Chemistry	Biology	...	Civic Education	A	B	C	A1	D1	A-BK	A1-BK	Highest_combi	year
0	0	1000001	1	1.0	2.2	3.50	NaN	NaN	NaN	NaN	...	NaN	NaN	NaN	11.50	NaN	NaN	NaN	NaN	C	2021
1	1	1000002	1	1.0	9.2	8.25	10.0	NaN	NaN	NaN	...	9.25	NaN	NaN	23.75	NaN	27.45	NaN	NaN	D1	2021
2	2	1000003	1	1.0	4.4	6.25	NaN	NaN	NaN	NaN	...	NaN	NaN	NaN	14.50	NaN	NaN	NaN	NaN	C	2021
3	3	1000004	1	1.0	8.4	6.75	10.0	4.00	3.00	3.50	...	NaN	15.40	14.90	NaN	22.40	25.15	17.8500	23.1000	D1	2021
4	4	1000005	1	1.0	8.8	8.25	9.4	8.25	5.75	5.25	...	NaN	22.80	19.80	NaN	26.45	26.45	23.7000	26.4375	D1	2021
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
921902	925728	64006212	64	6.0	5.4	8.25	5.0	NaN	NaN	NaN	...	7.75	NaN	NaN	18.75	NaN	18.65	NaN	NaN	C	2018
921903	925729	64006213	64	6.0	6.8	8.75	5.0	5.00	4.50	3.75	...	NaN	16.30	15.05	NaN	16.80	20.55	17.3250	17.7000	D1	2018
921904	925730	64006214	64	6.0	6.2	8.25	3.8	6.00	6.25	6.00	...	NaN	18.45	18.45	NaN	16.00	18.25	18.4875	16.6500	A	2018
921905	925731	64006215	64	6.0	6.8	7.00	2.2	6.25	6.75	4.00	...	NaN	19.80	17.55	NaN	15.25	16.00	19.9500	16.5375	A	2018
921906	925732	64006216	64	6.0	6.2	8.75	5.2	NaN	NaN	NaN	...	8.00	NaN	NaN	17.75	NaN	20.15	NaN	NaN	D1	2018

3689990 rows × 22 columns

Participant number year to year

def count_year():
    A = bigData['year'].value_counts().sort_index()
    CountYear = {'year':pd.Series(A.index.astype('str')),
                 'Count':pd.Series(A.values)}
    df = pd.DataFrame(CountYear)
    fig = px.bar(df, x='year', y='Count', labels={'Count':'Participant number'})
    return fig

count_year()

Stacked Area Chart

def count_range(data,subject,mini,maxi):
    df1 = data[data[subject] >= mini]
    df2 = df1[df1[subject] <= maxi]
    return len(df2)

import plotly.graph_objects as go

def AreaStack(subject):
    BigData = {2018: result(2018),
               2019: result(2019),
               2020: result(2020),
               2021: result(2021)}

    r1 = [count_range(BigData[2018+i],subject,0,5.1) for i in range(4)]
    r2 = [count_range(BigData[2018+i],subject,5.2,7.1) for i in range(4)]
    r3 = [count_range(BigData[2018+i],subject,7.2,8.0) for i in range(4)]
    r4 = [count_range(BigData[2018+i],subject,8.1,10) for i in range(4)]

    x=[2018,2019,2020,2021]
    fig = go.Figure()

    fig.add_trace(go.Scatter(
        x=x, y=r1,
        mode='lines',
        line=dict(width=0.5, color='rgb(190, 87, 12)'),
        stackgroup='one',
        name = '0 - 5.1',
        groupnorm='percent' # sets the normalization for the sum of the stackgroup
    ))
    fig.add_trace(go.Scatter(
        x=x, y=r2,
        mode='lines',
        line=dict(width=0.5, color='rgb(225, 105, 180)'),
        name = '5.2 - 7.1',
        stackgroup='one'
    ))
    fig.add_trace(go.Scatter(
        x=x, y=r3,
        mode='lines',
        line=dict(width=0.5, color='rgb(255, 220, 70)'),
        name = '7.2 - 8.0',
        stackgroup='one'
    ))
    fig.add_trace(go.Scatter(
        x=x, y=r4,
        mode='lines',
        line=dict(width=0.5, color='rgb(255, 10, 40)'),
        name = '8.1 - 10',
        stackgroup='one'
    ))

    fig.update_layout(
        showlegend=True,
        xaxis_type='category',
        yaxis=dict(
            type='linear',
            range=[1, 100],
            ticksuffix='%'),
        title = subject,
        legend_title = 'Score_range')

    fig.show()

for i in data.columns[4:13]:
    AreaStack(i)

Clustering province based on English distribution

D = data[data['CityCode']==60]
E = D[D['English'] >= 0]['English']

Left = []
Right = []
Med = []
for i in City:
    D = data[data['CityCode']==i]
    E = D[D['English'] >= 0]['English']
    if stats.skew(E) > 0.2:
        Left.append(i)
    elif stats.skew(E) < -0.2:
        Right.append(i)
    else:
        Med.append(i)
print(Left)
print(Right)
print(Med)

[5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 18, 22, 23, 28, 29, 30, 31, 32, 33, 34, 35, 36, 38, 39, 40, 43, 45, 46, 49, 50, 54, 56, 58, 59, 61, 62, 63, 64]
[1, 2, 3, 4, 44, 52]
[16, 17, 19, 21, 24, 25, 26, 27, 37, 41, 42, 47, 48, 51, 53, 55, 57, 60]

Correlation

sns.scatterplot(x = 'Math',y = 'Physics',data = data)
plt.rcParams['figure.figsize'] = (8,8)
plt.plot([0,10],[0,10],'r')
plt.show()

M = data[data['Math']>=0]
P = M[M['Physics']>=0]
Q = P[P['Physics']>=8.5]

print(stats.pearsonr(P['Math'],P['Physics']))
print(stats.spearmanr(P['Math'],P['Physics']))

(0.5943609804295731, 0.0)
SpearmanrResult(correlation=0.5882727656665806, pvalue=0.0)

sns.scatterplot(x = 'Math',y = 'Physics',data = P)
plt.rcParams['figure.figsize'] = (8,8)
plt.plot([0,10],[0,10],'r')
plt.show()

Hypothesis testing

Sample: Math score of 100 participants whose Physics score is greater or equal 8.

# Score spectrum of Math score of participants whose Physics score is greater or equal 8.
score_spectrum(data[data['Math']>=9],'Physics')

def MeanStd(data, subject):
    res = []
    for i in data[subject]:
        if i>=0:
            res.append(i)
    mean = sta.mean(res)
    stdev = sta.stdev(res)
    return [mean, stdev]

# Mean, std of data 
MeanStd(data[data['Math']>=9],'Physics')

[7.727144604187795, 1.091695307484447]

# Sample of 100 participant.

df100 = data[data['Physics']>=8].sample(1000)
score_spectrum(df100,'Math').show()
print(MeanStd(df100,'Math'))

[8.567, 0.5202515369142716]

def P_testing(df, subject1, subject2, ny0, alpha = 0.05, min_score = 8, n = 100):

    dfN = df[df[subject1]>=min_score].sample(n)
    
    mean = MeanStd(dfN, subject2)[0]
    s = MeanStd(dfN, subject2)[1]

    zc = (mean - ny0)/(s/n**0.5)
    p_value = 2*norm.cdf(zc)
    
    if zc > 0:
        p_value = 2*norm.cdf(-zc)
    else:
        p_value = 2*norm.cdf(zc)
    
    print(mean)
    print(s)
    print('====== HYPOTHESIS ======')
    print('H0: mean =',ny0)
    print('H1: mean !=',ny0)
    print('******** RESULT ********')
    if p_value < alpha:
        print('Reject H0')
    else:
        print('Accept H0')
    print('p-value:', p_value)

P_testing(df = data, subject1='Physics', subject2='Math', ny0=8.6, alpha=0.05, min_score=8, n=1000)

8.5742
0.5251000762617447
====== HYPOTHESIS ======
H0: mean = 8.6
H1: mean != 8.6
******** RESULT ********
Accept H0
p-value: 0.12024706393525411

Choropleth map

def Choropleth_VN_2020(feature):

    ProvinceInfo = pd.read_excel('data/ProvinceInfo.xlsx')
    #Vietnam map
    vietnam_geo = json.load(open("data/vietnam_state.geojson","r"))

    # Plotting
    fig = px.choropleth_mapbox(
        data_frame = ProvinceInfo,
        locations = 'CityCode',
        featureidkey='properties.Code',
        geojson = vietnam_geo,
        color = feature,
        hover_name = "CityName",
        mapbox_style = "carto-positron",
        center = {"lat": 16,"lon": 106},
        zoom = 4.5,
        title = "Vietnam Economy and Social map",
        color_continuous_scale=px.colors.diverging.Tealrose,

        width= 800,
        height=1000)
    fig.update_geos(fitbounds = "locations", visible=False)
    return fig

ProvinceInfodf = pd.read_excel('data/ProvinceInfo.xlsx')
ProvinceInfodf

	CityName	CityCode	Area	Monthly Income	Province Specialized Schools	Ethnic Pupils Ratio (%)	Students/ Teacher	Urban Ratio	Poverty Rate	Covid 19
0	Hà Nội	1	1	6205.45	8	2.12	21.93	49.25	0.4904	14421
1	Vĩnh Phúc	16	1	4289.92	1	10.48	21.73	29.97	2.4838	2862
2	Bắc Ninh	19	1	5439.20	1	4.39	15.20	31.47	0.7058	46267
3	Quảng Ninh	17	1	4539.47	1	0.16	19.82	64.44	2.1335	81
4	Hải Dương	21	1	4388.45	1	0.33	21.13	32.22	1.8446	1579
...	...	...	...	...	...	...	...	...	...	...
58	Cần Thơ	55	6	5031.09	1	3.38	17.63	70.02	1.9760	13
59	Hậu Giang	64	6	3974.29	1	3.08	19.38	27.97	5.5990	1
60	Sóc Trăng	59	6	3635.02	1	27.15	16.29	32.33	4.9784	4
61	Bạc Liêu	60	6	3338.48	1	7.79	23.37	27.88	6.6718	51
62	Cà Mau	61	6	3034.40	1	2.41	18.36	22.71	5.9179	13

63 rows × 10 columns

for i in ProvinceInfodf.columns[2:]:
    Choropleth_VN_2020(i).show()